|
{ |
|
"_name_or_path": "KoichiYasuoka/bert-base-russian-upos", |
|
"architectures": [ |
|
"BertForTokenClassification" |
|
], |
|
"attention_probs_dropout_prob": 0.1, |
|
"classifier_dropout": null, |
|
"directionality": "bidi", |
|
"hidden_act": "gelu", |
|
"hidden_dropout_prob": 0.1, |
|
"hidden_size": 768, |
|
"id2label": { |
|
"0": "LABEL_0", |
|
"1": "LABEL_1", |
|
"2": "LABEL_2", |
|
"3": "LABEL_3", |
|
"4": "LABEL_4", |
|
"5": "LABEL_5" |
|
}, |
|
"initializer_range": 0.02, |
|
"intermediate_size": 3072, |
|
"label2id": { |
|
"LABEL_0": 0, |
|
"LABEL_1": 1, |
|
"LABEL_2": 2, |
|
"LABEL_3": 3, |
|
"LABEL_4": 4, |
|
"LABEL_5": 5 |
|
}, |
|
"layer_norm_eps": 1e-12, |
|
"max_position_embeddings": 512, |
|
"model_type": "bert", |
|
"num_attention_heads": 12, |
|
"num_hidden_layers": 12, |
|
"output_past": true, |
|
"pad_token_id": 0, |
|
"pooler_fc_size": 768, |
|
"pooler_num_attention_heads": 12, |
|
"pooler_num_fc_layers": 3, |
|
"pooler_size_per_head": 128, |
|
"pooler_type": "first_token_transform", |
|
"position_embedding_type": "absolute", |
|
"task_specific_params": { |
|
"upos_multiword": { |
|
"ADJ+ADJ": { |
|
"\u0432\u043e\u0435\u043d\u043d\u043e\u043c\u043e\u0440\u0441\u043a\u0430\u044f": [ |
|
"\u0432\u043e\u0435\u043d\u043d\u043e", |
|
"\u043c\u043e\u0440\u0441\u043a\u0430\u044f" |
|
], |
|
"\u0432\u043e\u0435\u043d\u043d\u043e\u043c\u043e\u0440\u0441\u043a\u0443\u044e": [ |
|
"\u0432\u043e\u0435\u043d\u043d\u043e", |
|
"\u043c\u043e\u0440\u0441\u043a\u0443\u044e" |
|
], |
|
"\u0433\u043e\u0440\u043d\u043e\u043c\u0435\u0442\u0430\u043b\u043b\u0443\u0440\u0433\u0438\u0447\u0435\u0441\u043a\u0438\u0439": [ |
|
"\u0433\u043e\u0440\u043d\u043e", |
|
"\u043c\u0435\u0442\u0430\u043b\u043b\u0443\u0440\u0433\u0438\u0447\u0435\u0441\u043a\u0438\u0439" |
|
], |
|
"\u0434\u043e\u043b\u0433\u043e\u0441\u0440\u043e\u0447\u043d\u044b\u043c": [ |
|
"\u0434\u043e\u043b\u0433\u043e", |
|
"\u0441\u0440\u043e\u0447\u043d\u044b\u043c" |
|
], |
|
"\u043a\u043e\u0441\u043c\u043e\u0437\u0435\u043c\u043d\u044b\u0445": [ |
|
"\u043a\u043e\u0441\u043c\u043e", |
|
"\u0437\u0435\u043c\u043d\u044b\u0445" |
|
], |
|
"\u0441\u0443\u0434\u0435\u0431\u043d\u043e\u043c\u0435\u0434\u0438\u0446\u0438\u043d\u0441\u043a\u043e\u0439": [ |
|
"\u0441\u0443\u0434\u0435\u0431\u043d\u043e", |
|
"\u043c\u0435\u0434\u0438\u0446\u0438\u043d\u0441\u043a\u043e\u0439" |
|
] |
|
}, |
|
"ADJ+NOUN": { |
|
"\u0420\u043e\u0441\u0433\u043e\u0441\u0441\u0442\u0440\u0430\u0445\u0430": [ |
|
"\u0420\u043e\u0441", |
|
"\u0433\u043e\u0441\u0441\u0442\u0440\u0430\u0445\u0430" |
|
], |
|
"\u0420\u043e\u0441\u043a\u0443\u043b\u044c\u0442\u0443\u0440\u044b": [ |
|
"\u0420\u043e\u0441", |
|
"\u043a\u0443\u043b\u044c\u0442\u0443\u0440\u044b" |
|
], |
|
"\u0433\u043e\u0441\u0437\u0430\u043a\u0443\u043f\u043e\u043a": [ |
|
"\u0433\u043e\u0441", |
|
"\u0437\u0430\u043a\u0443\u043f\u043e\u043a" |
|
], |
|
"\u0433\u043e\u0441\u043a\u043e\u0440\u043f\u043e\u0440\u0430\u0446\u0438\u044f\u043c": [ |
|
"\u0433\u043e\u0441", |
|
"\u043a\u043e\u0440\u043f\u043e\u0440\u0430\u0446\u0438\u044f\u043c" |
|
], |
|
"\u0433\u043e\u0441\u043f\u043e\u0434\u0434\u0435\u0440\u0436\u043a\u0430": [ |
|
"\u0433\u043e\u0441", |
|
"\u043f\u043e\u0434\u0434\u0435\u0440\u0436\u043a\u0430" |
|
], |
|
"\u0433\u043e\u0441\u0441\u043b\u0443\u0436\u0431\u044b": [ |
|
"\u0433\u043e\u0441", |
|
"\u0441\u043b\u0443\u0436\u0431\u044b" |
|
], |
|
"\u043c\u0435\u0434\u0443\u0447\u0440\u0435\u0436\u0434\u0435\u043d\u0438\u0438": [ |
|
"\u043c\u0435\u0434", |
|
"\u0443\u0447\u0440\u0435\u0436\u0434\u0435\u043d\u0438\u0438" |
|
], |
|
"\u0441\u0435\u043b\u044c\u0445\u043e\u0437\u043f\u0440\u043e\u0434\u0443\u043a\u0446\u0438\u0438": [ |
|
"\u0441\u0435\u043b\u044c\u0445\u043e\u0437", |
|
"\u043f\u0440\u043e\u0434\u0443\u043a\u0446\u0438\u0438" |
|
] |
|
}, |
|
"ADJ+VERB": { |
|
"\u041b\u0435\u0432\u043e\u043d\u0430\u0441\u0442\u0440\u043e\u0435\u043d\u043d\u044b\u0435": [ |
|
"\u041b\u0435\u0432\u043e", |
|
"\u043d\u0430\u0441\u0442\u0440\u043e\u0435\u043d\u043d\u044b\u0435" |
|
] |
|
}, |
|
"ADP+DET": { |
|
"\u0441\u0434\u0440\u0443\u0433\u0438\u043c\u0438": [ |
|
"\u0441", |
|
"\u0434\u0440\u0443\u0433\u0438\u043c\u0438" |
|
], |
|
"\u0441\u0442\u0435\u0445": [ |
|
"\u0441", |
|
"\u0442\u0435\u0445" |
|
] |
|
}, |
|
"ADP+PRON": { |
|
"\u0441\u043d\u0438\u043c": [ |
|
"\u0441", |
|
"\u043d\u0438\u043c" |
|
] |
|
}, |
|
"ADV+VERB": { |
|
"\u043d\u0435\u0433\u0434\u0435\u0442\u0440\u0435\u043d\u0438\u0440\u043e\u0432\u0430\u0442\u044c": [ |
|
"\u043d\u0435\u0433\u0434\u0435", |
|
"\u0442\u0440\u0435\u043d\u0438\u0440\u043e\u0432\u0430\u0442\u044c" |
|
] |
|
}, |
|
"NUM+NOUN": { |
|
"\u041f\u043e\u043b\u043b\u0438\u0442\u0440\u0430": [ |
|
"\u041f\u043e\u043b", |
|
"\u043b\u0438\u0442\u0440\u0430" |
|
], |
|
"\u043f\u043e\u043b\u0432\u0435\u043a\u0430": [ |
|
"\u043f\u043e\u043b", |
|
"\u0432\u0435\u043a\u0430" |
|
], |
|
"\u043f\u043e\u043b\u0433\u043e\u0434\u0430": [ |
|
"\u043f\u043e\u043b", |
|
"\u0433\u043e\u0434\u0430" |
|
], |
|
"\u043f\u043e\u043b\u0434\u043d\u044f": [ |
|
"\u043f\u043e\u043b", |
|
"\u0434\u043d\u044f" |
|
], |
|
"\u043f\u043e\u043b\u0434\u044e\u0436\u0438\u043d\u044b": [ |
|
"\u043f\u043e\u043b", |
|
"\u0434\u044e\u0436\u0438\u043d\u044b" |
|
], |
|
"\u043f\u043e\u043b\u043a\u0438\u043b\u043e": [ |
|
"\u043f\u043e\u043b", |
|
"\u043a\u0438\u043b\u043e" |
|
], |
|
"\u043f\u043e\u043b\u043a\u0438\u043b\u043e\u0433\u0440\u0430\u043c\u043c\u0430": [ |
|
"\u043f\u043e\u043b", |
|
"\u043a\u0438\u043b\u043e\u0433\u0440\u0430\u043c\u043c\u0430" |
|
], |
|
"\u043f\u043e\u043b\u043a\u043e\u0442\u043b\u0435\u0442\u044b": [ |
|
"\u043f\u043e\u043b", |
|
"\u043a\u043e\u0442\u043b\u0435\u0442\u044b" |
|
], |
|
"\u043f\u043e\u043b\u043a\u0443\u0441\u043e\u0447\u043a\u0430": [ |
|
"\u043f\u043e\u043b", |
|
"\u043a\u0443\u0441\u043e\u0447\u043a\u0430" |
|
], |
|
"\u043f\u043e\u043b\u043c\u0435\u0441\u044f\u0446\u0430": [ |
|
"\u043f\u043e\u043b", |
|
"\u043c\u0435\u0441\u044f\u0446\u0430" |
|
], |
|
"\u043f\u043e\u043b\u043c\u0435\u0442\u0440\u0430": [ |
|
"\u043f\u043e\u043b", |
|
"\u043c\u0435\u0442\u0440\u0430" |
|
], |
|
"\u043f\u043e\u043b\u043c\u0435\u0448\u043a\u0430": [ |
|
"\u043f\u043e\u043b", |
|
"\u043c\u0435\u0448\u043a\u0430" |
|
], |
|
"\u043f\u043e\u043b\u043c\u0438\u043b\u043b\u0438\u0430\u0440\u0434\u0430": [ |
|
"\u043f\u043e\u043b", |
|
"\u043c\u0438\u043b\u043b\u0438\u0430\u0440\u0434\u0430" |
|
], |
|
"\u043f\u043e\u043b\u043c\u0438\u043b\u043b\u0438\u043e\u043d\u0430": [ |
|
"\u043f\u043e\u043b", |
|
"\u043c\u0438\u043b\u043b\u0438\u043e\u043d\u0430" |
|
], |
|
"\u043f\u043e\u043b\u043c\u0438\u043d\u0443\u0442\u044b": [ |
|
"\u043f\u043e\u043b", |
|
"\u043c\u0438\u043d\u0443\u0442\u044b" |
|
], |
|
"\u043f\u043e\u043b\u043c\u044b\u0441\u043b\u0438": [ |
|
"\u043f\u043e\u043b", |
|
"\u043c\u044b\u0441\u043b\u0438" |
|
], |
|
"\u043f\u043e\u043b\u043f\u0443\u0442\u0438": [ |
|
"\u043f\u043e\u043b", |
|
"\u043f\u0443\u0442\u0438" |
|
], |
|
"\u043f\u043e\u043b\u0447\u0430\u0441\u0430": [ |
|
"\u043f\u043e\u043b", |
|
"\u0447\u0430\u0441\u0430" |
|
] |
|
}, |
|
"NUM+SYM": { |
|
"20\u00b0": [ |
|
"20", |
|
"\u00b0" |
|
], |
|
"2\u00b0": [ |
|
"2", |
|
"\u00b0" |
|
], |
|
"90\u00b0": [ |
|
"90", |
|
"\u00b0" |
|
] |
|
}, |
|
"PART+PRON": { |
|
"\u0434\u0430\u0442\u043e": [ |
|
"\u0434\u0430", |
|
"\u0442\u043e" |
|
] |
|
}, |
|
"SYM+NUM": { |
|
"\u21161": [ |
|
"\u2116", |
|
"1" |
|
], |
|
"\u2116159": [ |
|
"\u2116", |
|
"159" |
|
], |
|
"\u211619": [ |
|
"\u2116", |
|
"19" |
|
], |
|
"\u21162": [ |
|
"\u2116", |
|
"2" |
|
], |
|
"\u211636": [ |
|
"\u2116", |
|
"36" |
|
], |
|
"\u21165": [ |
|
"\u2116", |
|
"5" |
|
] |
|
}, |
|
"VERB+ADV": { |
|
"\u041d\u0435\u043a\u043e\u0433\u0434\u0430": [ |
|
"\u041d\u0435", |
|
"\u043a\u043e\u0433\u0434\u0430" |
|
], |
|
"\u043d\u0435\u0437\u0430\u0447\u0435\u043c": [ |
|
"\u043d\u0435", |
|
"\u0437\u0430\u0447\u0435\u043c" |
|
], |
|
"\u043d\u0435\u043a\u043e\u0433\u0434\u0430": [ |
|
"\u043d\u0435", |
|
"\u043a\u043e\u0433\u0434\u0430" |
|
], |
|
"\u043d\u0435\u043a\u0443\u0434\u0430": [ |
|
"\u043d\u0435", |
|
"\u043a\u0443\u0434\u0430" |
|
] |
|
}, |
|
"VERB+PRON": { |
|
"\u041d\u0435\u043a\u043e\u043c\u0443": [ |
|
"\u041d\u0435", |
|
"\u043a\u043e\u043c\u0443" |
|
], |
|
"\u043d\u0435\u043a\u043e\u0433\u043e": [ |
|
"\u043d\u0435", |
|
"\u043a\u043e\u0433\u043e" |
|
], |
|
"\u043d\u0435\u043a\u043e\u043c\u0443": [ |
|
"\u043d\u0435", |
|
"\u043a\u043e\u043c\u0443" |
|
], |
|
"\u043d\u0435\u0447\u0435\u0433\u043e": [ |
|
"\u043d\u0435", |
|
"\u0447\u0435\u0433\u043e" |
|
], |
|
"\u043d\u0435\u0447\u0435\u043c": [ |
|
"\u043d\u0435", |
|
"\u0447\u0435\u043c" |
|
], |
|
"\u043d\u0435\u0447\u0435\u043c\u0443": [ |
|
"\u043d\u0435", |
|
"\u0447\u0435\u043c\u0443" |
|
] |
|
}, |
|
"X+X": { |
|
"xxx": [ |
|
"x", |
|
"xx" |
|
], |
|
"xxxx": [ |
|
"x", |
|
"xxx" |
|
] |
|
} |
|
} |
|
}, |
|
"tokenizer_class": "BertTokenizerFast", |
|
"torch_dtype": "float32", |
|
"transformers_version": "4.24.0", |
|
"type_vocab_size": 2, |
|
"use_cache": true, |
|
"vocab_size": 119547 |
|
} |
|
|